[Eeglablist] Downsizing Number of Electrodes

[Makoto Miyakoshi]

Dear Marco,

>> My concern is whether the data would
>> look different in the two scenarios i.e. working on 32 electrodes, would,
>> let's say, data from electrode Cz look different from data from electrode Cz
>> obtained from working with, initially, 10 electrodes only?

Unless you do ICA or average referencing, no. It does not change your
preprocess results. That said, I agree with Steve. Reducing the number
of channels from 32 to 10 does not reduce the amount of your effort
required for preprocesses.

Makoto

2012/6/24 Stephen Politzer-Ahles <politzerahless at gmail.com>:
> Hi Marco,
>
> Other than the size of the datasets, I can't think of any drawbacks to
> working with all 32 electrodes from the beginning. I'm not sure what exactly
> your data processing entails, but at least in my experience most of the data
> processing steps (e.g. filtering, baseline correction, epoching) work
> exactly the same whether you have many or few electrodes. For processes
> (such as filtering and baseline correction), electrodes should be
> independent from one another, but for some things (such as ICA
> decomposition, and possibly referencing--if you want to use average
> reference) the number of electrodes does matter, and in every case I can
> think of it's best to have as much of your data as possible. So my thought
> would be, as long as you have sufficient space, it's probably best to keep
> all electrodes through the entire data processing.
>
> Best,
> Steve
>
> On Sat, Jun 23, 2012 at 10:46 PM, Marco Montalto <montaltomarco at onvol.net>
> wrote:
>>
>> Dear List,
>>
>> I have what most probably is a really banal question but which I thought I
>> should gather advise from the experts regarding the issue before committing
>> myself to a decision. I am collecting data from 32 electrodes. Now at the
>> final stages I might want to use data from only 10 of those electrodes for
>> statistical analysis. From your experience do you think running the entire
>> data processing procedure on the 32 electrodes and then running statistical
>> analysis on only 10 of those electrodes would be a problem? I envisage that
>> working on 10 electrodes only from the very start would require much less
>> work but I don't mind the added task. My concern is whether the data would
>> look different in the two scenarios i.e. working on 32 electrodes, would,
>> let's say, data from electrode Cz look different from data from electrode Cz
>> obtained from working with, initially, 10 electrodes only?
>>
>> Looking forward to your replies.
>>
>> Thanks and regards,
>> Marco
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>
>
>
>
> --
> Stephen Politzer-Ahles
> University of Kansas
> Linguistics Department
> http://www.linguistics.ku.edu/
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu



-- 
Makoto Miyakoshi
JSPS Postdoctral Fellow for Research Abroad
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego